CN-122021233-A - Method for improving sealing performance of fluorine lining conduit based on intelligent optimization algorithm

CN122021233ACN 122021233 ACN122021233 ACN 122021233ACN-122021233-A

Abstract

The invention discloses a sealing performance improving method of a fluorine lining conduit based on an intelligent optimization algorithm, which comprises the following steps of collecting relevant data of the fluorine lining conduit and carrying out pretreatment to generate a multi-dimensional input data set, establishing a mixed digital twin model to obtain a sealing performance prediction result, constructing an improved water circulation algorithm optimization framework to generate an optimization control parameter, optimizing the framework based on the improved water circulation algorithm to form a searching updating result, obtaining a candidate parameter sealing performance result and a next generation candidate parameter set by utilizing the mixed digital twin model, applying the optimal parameter set to manufacturing and assembling processes of the fluorine lining conduit to generate a manufacturing and assembling result, and obtaining an updated optimal parameter set based on the manufacturing and assembling result and data acquired in real time. The invention adopts a mixed digital twin and improved water circulation algorithm to realize dynamic optimization of sealing performance of the fluorine lining conduit.

Inventors

HAN YUXIN
WANG ZHENGDONG
FAN GUOHUA

Assignees

泰州市三星铁氟隆有限公司

Dates

Publication Date: 20260512
Application Date: 20251204

Claims (8)

1. The method for improving the sealing performance of the fluorine lining catheter based on the intelligent optimization algorithm is characterized by comprising the following steps of: collecting and preprocessing relevant data of the fluorine lining catheter to generate a multi-dimensional input data set; establishing a mixed digital twin model based on a multi-dimensional input data set, training and updating in real time to obtain a sealing performance prediction result; An improved water circulation algorithm optimization framework is constructed based on the sealing performance prediction result, a multi-objective optimization function is set, a dynamic weight decomposition strategy is introduced, and an optimization control parameter is generated; in an improved water circulation algorithm optimization framework, guiding rainfall and evaporation search strategies based on a sealing performance prediction result to form a search update result; Initializing a candidate parameter set by utilizing the optimized control parameters and the search updating results, and inputting a mixed digital twin model to obtain a candidate parameter sealing performance result and a next generation candidate parameter set; judging an optimal parameter set based on the candidate parameter sealing performance result, and applying the optimal parameter set to the manufacturing and assembling processes of the fluorine lining conduit to generate a manufacturing and assembling result; And collecting real-time data in the production operation stage, inputting the real-time data and the manufacturing assembly result into a mixed digital twin model, and outputting an updated optimal parameter set.
2. The method for improving the sealing performance of the fluorine lining catheter based on the intelligent optimization algorithm according to claim 1, wherein the preprocessing comprises dimension unification, outlier rejection, time stamp marking and multidimensional feature mapping.
3. The method for improving the sealing performance of the fluorine lining conduit based on the intelligent optimization algorithm according to claim 1, wherein the method for improving the sealing performance of the fluorine lining conduit based on the multi-dimensional input data set is characterized by comprising the following steps of: dividing a multi-dimensional input data set into a historical test data set and an online sensing data set, establishing a finite element contact mechanical model and a fluid-solid coupling model on the basis of the historical test data set, and calculating the stress and fluid pressure change of the fluorine lining conduit under different working conditions to obtain a physical mechanism feature matrix; Based on the physical mechanism feature matrix and the historical test data set, constructing a deep neural network model, setting a loss function, and continuously iterating and updating network parameters through a supervised learning method to obtain deep neural network model parameters; inputting the online sensing data set and the initial neural network model parameters into a deep neural network model, and carrying out gradual gradient update on the network parameters by combining the real-time calculation results of the finite element contact mechanical model and the fluid-solid coupling model to form a real-time updated hybrid digital twin model; And (3) carrying out reasoning calculation on the multi-dimensional input data set by utilizing a real-time updated mixed digital twin model, and outputting a sealing performance prediction result set comprising a predicted leakage rate, a predicted critical failure pressure, an uncertainty and a predicted fatigue life.
4. The method for improving the sealing performance of the fluorine lining conduit based on the intelligent optimization algorithm according to claim 1, wherein the steps of constructing an improved water circulation algorithm optimization framework based on the sealing performance prediction result, setting a multi-objective optimization function, introducing a dynamic weight decomposition strategy, and generating optimization control parameters specifically comprise: Based on a sealing performance prediction result set, constructing a multi-objective optimization function comprising four objectives of leakage rate, critical failure pressure, fatigue life and manufacturing cost, and setting weight coefficients for the four objectives respectively; Establishing a constraint condition set based on a multi-objective optimization function, setting an assembly moment range, a fluoroplastic material strength limit and a manufacturing tolerance requirement as inequality constraint conditions, and setting a geometric relationship and physical consistency of a sealing structure as equality constraint conditions; Constructing a self-adaptive penalty function on the basis of a constraint condition set, introducing a penalty coefficient dynamically adjusted along with the iteration step number into a comprehensive optimization target, applying a penalty value in a square form to a candidate design scheme violating the constraint, and adding the penalty value with a multi-target optimization function to form the comprehensive optimization target function; In an improved water circulation algorithm framework, generating an initial candidate design scheme parameter set according to a multi-objective optimization function, a constraint condition set and a comprehensive optimization objective function, and treating the initial candidate design scheme parameter set as a water drop set; Calculating an initial comprehensive optimization target value of each candidate design scheme, and dividing the candidate design schemes into river sets and stream sets according to the initial comprehensive optimization target value; In the iteration process of the improved water circulation algorithm, taking uncertainty in a sealing performance prediction result set as a guide, executing global dominant search on candidate design schemes in a river set, executing local follow-up search on the candidate design schemes in a stream set, triggering evaporation and rainfall operation on the candidate design schemes with lower comprehensive optimization target values, and generating new candidate design schemes; Continuously executing the operations of the collection of the river set, the migration of the stream set and the generation of new candidate design schemes by evaporation and rainfall, and ending the iterative process and outputting the optimized control parameters when the difference value of the comprehensive optimized target values of two continuous iterations is smaller than a preset convergence threshold value or reaches the set maximum iteration times.
5. The method for improving the sealing performance of a fluorine lining pipe based on an intelligent optimization algorithm according to claim 1, wherein in the improved water circulation algorithm optimization framework, guiding a rainfall and evaporation search strategy based on a sealing performance prediction result, and forming a search update result specifically comprises: In an improved water circulation algorithm optimization framework, uncertainty of a sealing performance prediction result set is used as real-time searching guide, rainfall operation is carried out on a parameter space region with uncertainty, a new candidate design scheme is generated, evaporation operation is carried out on the parameter space region with uncertainty, and local searching of the existing candidate design scheme is enhanced; Calculating the contact stress gradient and the thermal expansion coefficient of each candidate design scheme at the current iteration moment, and dynamically correcting the original iteration step length to obtain a corrected step length factor; Carrying out iterative updating on the parameter vector of each candidate design scheme by utilizing the corrected step size factor, and calculating to obtain the parameter vector of the next iterative step; After each iteration is finished, reevaluating and classifying all the candidate design schemes according to the latest comprehensive optimization target value and uncertainty; continuously executing uncertainty-guided rainfall and evaporation operation, step correction based on contact stress gradient and thermal expansion coefficient, and parameter iterative updating of candidate design schemes until the comprehensive optimization target meets a preset convergence condition or reaches a set maximum iteration number, and outputting a search updating result.
6. The method for improving the sealing performance of the fluorine lining conduit based on the intelligent optimization algorithm according to claim 1, wherein initializing a candidate parameter set by using the optimization control parameter and the search update result, inputting a hybrid digital twin model, and obtaining the candidate parameter sealing performance result and the next generation candidate parameter set specifically comprises: based on the optimized control parameters and the search updating results, randomly generating initial candidate parameter sets in a set parameter range, and distributing a unique number and an initial iteration step number for each candidate parameter set; inputting each initial candidate parameter set into a mixed digital twin model updated in real time, calculating a sealing performance predicted value of the candidate parameter set, and outputting a sealing performance predicted result; substituting the sealing performance prediction result of each candidate parameter set into a comprehensive optimization objective function, and calculating and recording a comprehensive optimization target value of each candidate parameter set under initial iteration by combining a multi-objective optimization function; classifying all candidate parameter sets according to the comprehensive optimization target value to obtain a river set and a stream set, executing evaporation operation on the candidate parameter set with the lowest comprehensive performance evaluation value and generating a new candidate parameter set; Combining the candidate parameter vector of the previous iteration step with the contact stress gradient, the thermal expansion coefficient and the uncertainty by utilizing the corrected step length factor in each iteration, calculating to obtain a new candidate parameter vector, and synchronously updating all candidate parameter sets in the river set and the stream set; and repeatedly executing iteration processes of sealing performance prediction, comprehensive optimization target calculation, hierarchical classification and parameter updating of the candidate parameter set until the variation of the comprehensive optimization target between two continuous iterations is smaller than a preset convergence threshold or reaches a set maximum iteration number, and outputting the finally formed next generation candidate parameter set.
7. The method for improving the sealing performance of the fluorine lining conduit based on the intelligent optimization algorithm according to claim 1, wherein the method for judging the optimal parameter set based on the candidate parameter sealing performance result is applied to the manufacturing and assembling process of the fluorine lining conduit, and the method for generating the manufacturing and assembling result specifically comprises the following steps: After each iteration is completed, inputting a candidate parameter sealing performance result obtained by the current iteration into a comprehensive optimization objective function, calculating a current comprehensive optimization objective value, and comparing the current comprehensive optimization objective value with a comprehensive optimization objective value corresponding to the previous iteration to obtain a comprehensive optimization objective value difference value; when the comprehensive optimization target value difference values of all the candidate parameter sets are smaller than a preset convergence accuracy threshold value or the iteration number reaches the set maximum iteration number, determining that the optimization process is converged; after the convergence of the optimization process is determined, selecting a candidate parameter set with the optimal comprehensive optimization target value from all candidate parameter sets to form an optimal parameter set; and applying the optimal parameter set to the manufacturing and assembling process of the fluorine lining conduit, and performing production operation to form a final manufacturing and assembling result.
8. The method for improving the sealing performance of the fluorine lining conduit based on the intelligent optimization algorithm according to claim 1, wherein the steps of collecting real-time data in the production operation stage, inputting the real-time data and the production assembly result into a hybrid digital twin model, and outputting the updated optimal parameter set specifically comprise: After the fluorine lining conduit enters the production operation stage, continuously acquiring operation monitoring data through sensors arranged at the pipe body and the connecting part; Inputting operation monitoring data and manufacturing and assembling results into a mixed digital twin model, and adaptively updating parameters and physical mechanism characteristics of a deep neural network part in the mixed digital twin model; the updated mixed digital twin model is utilized to recalculate the sealing performance, so that the latest sealing performance prediction result is obtained, and the latest uncertainty is evaluated; When the latest uncertainty exceeds a preset threshold, triggering an improved water circulation algorithm to enter a new round of optimization iteration; in the new optimization iteration process, grouping and updating the candidate parameter sets according to the latest uncertainty, and dynamically adjusting the parameter vectors of the candidate parameter sets; After iteration reaches a preset convergence condition or maximum iteration times, outputting an updated optimal parameter set, and replacing the original optimal parameter set to be applied to the production, the manufacture, the operation and the maintenance of the fluorine lining conduit.

Description

Method for improving sealing performance of fluorine lining conduit based on intelligent optimization algorithm Technical Field The invention relates to the technical field of sealing optimization of fluorine lining pipes, in particular to a method for improving sealing performance of fluorine lining pipes based on an intelligent optimization algorithm. Background The existing fluorine lining conduit sealing design and manufacture mainly depend on finite element analysis, experimental data and manual experience for parameter determination. The common method comprises the steps of seal structure simulation calculation based on finite element contact mechanics, multi-objective optimization design based on a response surface, and parameter optimization by adopting a genetic algorithm, a particle swarm algorithm or a standard water circulation algorithm. The method can improve the geometric dimension of the sealing groove, the compression ratio of the gasket and the rationality of heat treatment process parameters to a certain extent, and can be combined with experiments to verify and predict the leakage rate, critical failure pressure and fatigue life. However, the prior art generally has the problems of long calculation period, poor adaptability to working condition change and insufficient coupling between parameter optimization results and actual production environments. Finite element models and traditional data driven models often have difficulty in reflecting dynamic changes in the production assembly process and temperature and pressure fluctuations in service operation in real time, resulting in a decline in prediction accuracy over time. When the conventional genetic algorithm, particle swarm algorithm and standard water circulation algorithm process a design space which is multi-target and multi-constraint and contains uncertainty, the search efficiency and the convergence speed are limited, and the physical information constraint aiming at a seal failure mechanism is lacking, so that the situations of local optimum and low feasible solution proportion are easy to occur. Meanwhile, most of the prior art stays in a single optimization stage of a design and manufacturing link, lacks a real-time feedback mechanism of manufacturing assembly results and operation monitoring data, and cannot continuously correct a model and process parameters in a long-term operation process of equipment. Therefore, how to provide a method for improving the sealing performance of a fluorine lining catheter based on an intelligent optimization algorithm is a problem to be solved by those skilled in the art. Disclosure of Invention The invention aims to provide a method for improving sealing performance of a fluorine lining conduit based on an intelligent optimization algorithm, which comprehensively applies a mixed digital twin model and an improved water circulation algorithm, and performs high-precision prediction on leakage rate, critical failure pressure and fatigue life by acquiring multidimensional input data in real time and constructing a prediction model of physical mechanism and data driving fusion, and performs dynamic optimization and iterative update of candidate parameter sets under the guidance of uncertainty so as to realize full-flow self-adaptive optimization of geometric dimensions of a sealing groove, compression rate of a gasket and heat treatment process parameters. The method can continuously collect temperature, pressure and leakage monitoring data and feed back the model in the production operation process, realizes closed-loop control of design, manufacture, assembly and operation, and has the advantages of high prediction precision, high parameter optimization efficiency, strong long-term operation stability and good production consistency. According to the embodiment of the invention, the method for improving the sealing performance of the fluorine lining catheter based on the intelligent optimization algorithm comprises the following steps: collecting and preprocessing relevant data of the fluorine lining catheter to generate a multi-dimensional input data set; establishing a mixed digital twin model based on a multi-dimensional input data set, training and updating in real time to obtain a sealing performance prediction result; An improved water circulation algorithm optimization framework is constructed based on the sealing performance prediction result, a multi-objective optimization function is set, a dynamic weight decomposition strategy is introduced, and an optimization control parameter is generated; in an improved water circulation algorithm optimization framework, guiding rainfall and evaporation search strategies based on a sealing performance prediction result to form a search update result; Initializing a candidate parameter set by utilizing the optimized control parameters and the search updating results, and inputting a mixed digital twin model to obtain a candidate parameter seali